Blog Posts — March 2, 2018

The Markets May Have Gone Haywire, But Factors Have Not

by Melissa R. Brown, CFA

While the US equity market has experienced sharply increased risk based on daily returns, intraday high-low spreads have experienced even more volatility (see “The Increase in Volatility? Yup, It Was Big”). This heightened volatility has largely been a market event; style factors have barely blinked.

For both February and year-to-date, factor returns in most cases were well within bounds of expectations, based on Axioma’s medium-horizon model forecasts. Tables 1 and 2 below show returns for February and year-to-date for the major model style factors across the major regions. The region with the highest return for a given factor is highlighted in green, and the lowest return is highlighted in pink. Importantly, the factors for which the return fell outside a two-standard-deviation range are asterisked. Relatively few factor returns merited this designation and, for most of those that did, the return was in the expected direction, which is typically not viewed as an issue. For example, Momentum (Medium-Term) has continued the pattern from last year of having unusually positive performance across many regions, with higher-than-expected returns in a number of regions, clearly a positive for Momentum-based investors.

With these well-behaved returns, few factors have seen sharp increases in volatility, and some have even experienced a drop. The charts below show daily volatility year to date for US factors. Only Dividend Yield, Size, and Volatility have seen increases in risk, and even those had average-or-below medium-horizon volatility at the end of February (the long-term average is at the top of each chart).

A quick review of our “sample factor portfolios” (in which we tilt on a particular factor and target 3% tracking error with a once-a-month rebalancing[1]) also showed no increase in tracking error (and in some cases saw a decrease) between the end of January rebalancings and the end of February.

Finally, we also looked at US factor correlations, which have been little-changed this year. Overall, this suggests that the outside turmoil of gyrating equity markets and increases in total portfolio risk, which may reasonably have been expected to drive up active risk, may not have had a major impact (at least in the US) on style-factor based portfolios.

[1] See our recent paper “When it Comes to Momentum, Don’t Cramp My Style” for more detail with a focus on Momentum.